ACTA ICHTHYOLOGICA ET PISCATORIA (2015) 45 (4): 427–431 DOI: 10.3750/AIP2015.45.4.13

LENGTH–WEIGHT RELATIONS FOR 29 DEMERSAL FISHES CAUGHT BY SMALL OTTER TRAWL ON THE SOUTH-EASTERN COAST OF KOREA

Joo Myun PARK1* and Sung Hoi HUH2

1Department of Biological Sciences, Macquarie University, Sydney, Australia 2Department of Oceanography, Pukyong National University, Busan, Korea

Park J.M., Huh S.H. 2015. Length–weight relations for 29 demersal fi shes caught by small otter trawl on the south-eastern coast of Korea. Acta Ichthyol. Piscat. 45 (4): 427–431. Abstract. We present herewith the relations between total length and weight (LWRs) for 29 fi sh species from the southern coast of the East Sea (=Sea of Japan), Korea: Conger myriaster (Brevoort, 1856); Coilia nasus Temminck et Schlegel, 1846; Engraulis japonicus Temminck et Schlegel, 1846; Clupea pallasii pallasii Valenciennes, 1847; Coelorinchus multispinulosus Katayama, 1942; Lophius litulon (Jordan, 1902); Zeus faber Linnaeus, 1758; Paracentropogon rubripinnis (Temminck et Schlegel, 1843); Lepidotrigla guentheri Hilgendorf, 1879; Liparis tanakae (Gilbert et Burke, 1912); Acropoma japonicum Günther, 1859; Doederleinia berycoides (Hilgendorf, 1879); Jaydia lineata (Temminck et Schlegel, 1842); Ostorhinchus semilineatus (Temminck et Schlegel, 1842); Sillago sihama (Forsskål, 1775); Trachurus japonicus (Temminck et Schlegel, 1844); Pagrus major (Temminck et Schlegel, 1843); Pennahia argentata (Houttuyn, 1782); Zoarces gillii Jordan et Starks, 1905; Callionymus lunatus Temminck et Schlegel, 1845; Callionymus valenciennei Temminck et Schlegel, 1845; Am- blychaeturichthys hexanema (Bleeker, 1853); Amblychaeturichthys sciistius (Jordan et Snyder, 1901); Sphyraena pinguis Günther, 1874; Trichiurus lepturus Linnaeus, 1758; Psenopsis anomala (Temminck et Schlegel, 1844); Pseudorhombus pentophthalmus Günther, 1862; Pseudopleuronectes yokohamae (Günther, 1877); and Cyno- glossus robustus Günther, 1873. The LWRs for 12 species are estimated for the fi rst time and new maximum lengths for six species are reported herewith. All of the relations between total length and weight were signifi cant (all r2 > 0.938). The values of exponent b, estimated using simple linear least squares of log-transformed weight and length data, ranged from 2.730 to 3.440. Keywords: marine fi sh, fi sheries management, East Sea, Sea of Japan, LWR

The fi shes examined in the presently reported study habitats and/or regions (Gonçalves et al. 1997). Although were collected off the south-eastern coast of Korea. The LWRs of scorpionfi shes and righteye fl ounders inhabiting study area, in terms of its environmental and geographical south-eastern coast of Korea have been reported (Baeck characteristics, differs generally from other coastal marine et al. 2012, 2014), there are still insuffi cient informa- ecosystems in the east and south seas of Korea (Suh et tion of LWR parameters for Korean fi shes. This study al. 2001). It is a habitat for many commercially important reports LWRs for 29 fi sh species collected by demersal demersal fi sh species and has been renowned for being trawl from the south-eastern coast of Korea: Conger abundant in prey organisms (Kim unpublished**). How- myriaster (Brevoort, 1856); Coilia nasus Temminck et ever, little is known on the ecology of the fi shes living in Schlegel, 1846; Engraulis japonicus Temminck et Schle- the area. gel, 1846; Clupea pallasii pallasii Valenciennes, 1847; Length–weight relation (LWR) parameters (a, b) are Coelorinchus multispinulosus Katayama, 1942; Lophius useful in the assessment of fi sh stocks and populations litulon (Jordan, 1902); Zeus faber Linnaeus, 1758; (Ricker 1968, Froese 2006). LWRs are used to estimate the Paracentropogon rubripinnis (Temminck et Schlegel, mean weight for a given length group and convert length 1843); Lepidotrigla guentheri Hilgendorf, 1879; Liparis observations into weights to provide a measure of biomass tanakae (Gilbert et Burke, 1912); Acropoma japonicum (Froese 2006, Froese et al. 2011). LWRs also allow for life Günther, 1859; Doederleinia berycoides (Hilgendorf, history and morphological comparisons between different 1879); Jaydia lineata (Temminck et Schlegel, 1842); Os- fi sh species or between fi sh populations from different torhinchus semilineatus (Temminck et Schlegel, 1842);

* Correspondence: Dr Joo Myun Park, Department of Biological Sciences, Macquarie University, Sydney, NSW 2109, Australia; phone: +61 2 9850 8234; e-mail: (JMP) [email protected] or [email protected], (SHH) [email protected] ** Kim D.J. 1998. [Seasonal variation of species composition of demersal fi sh off Kori.] MSc Thesis, Pukyong National University, Korea. [In Korean.] 428 Park and Huh

Sillago sihama (Forsskål, 1775); Trachurus japonicus hama. All of b values were within the expected range of (Temminck et Schlegel, 1844); Pagrus major (Temminck 2.5 to 3.5 (Froese 2006). The mean value of b (± stan- et Schlegel, 1843); Pennahia argentata (Houttuyn, 1782); dard deviation) was 3.117 ± 0.184 indicating a tenden- Zoarces gillii Jordan et Starks, 1905; Callionymus lunatus cy towards positive allometry, which is followed by the Temminck et Schlegel, 1845; Callionymus valenciennei majority of fi sh species (Froese 2006). The LWRs for 12 Temminck et Schlegel, 1845; Amblychaeturichthys of the 29 species determined in the course of the pres- hexanema (Bleeker, 1853); Amblychaeturichthys sciisti- ently reported study are fi rst reports of LWRs for those us (Jordan et Snyder, 1901); Sphyraena pinguis Günther, species. For remaining 17 species, LWRs of 8 species, 1874; Trichiurus lepturus Linnaeus, 1758; Psenopsis Amblychaeturichthys hexanema, Jaydia lineata, Conger anomala (Temminck et Schlegel, 1844); Pseudorhombus myriaster, Engraulis japonicus, Liparis tanakae, Lophi- pentophthalmus Günther, 1862; Pseudopleuronectes us litulon, Pseudopleuronectes yokohamae, and S. siha- yokohamae (Günther, 1877); and Cynoglossus robustus ma have been reported from the Yellow Sea (Xue et al. Günther, 1873. 2011, Wang et al. 2013) and the southern coast of Korea Fish specimens were collected monthly from January (Baeck et al. 2014). The LWRs of 15 species also were to December 2006 off the south-eastern coast of Korea available in FishBase (Froese and Pauly 2015), and 2 (35°15′~35°17′N, 129°18′~129°21′E) from depths be- species, A. hexanema and L. litulon were only listed in tween 30 and 60 m using a small otter trawl. Fish samples the literature (Xue et al. 2011, Wang et al. 2013). were immediately packed with ice and taken to the labora- The seasonal LWRs for 10 of the 29 species are present- tory for measurement of total length (TL) and wet weight ed in Table 2. The parameter b did not differ signifi cant- to the nearest millimetre and gram, respectively. All scien- ly (ANCOVA; P > 0.05) with seasons for Jaydia lineata, tifi c names were verifi ed following FishBase (Froese and Coelorinchus multispinulosus, Lophius litulon, Trachurus Pauly 2015). japonicus, Callionymus lunatus, and Pseudopleuronectes For each species, the weight–length relation was deter- yokohamae, whereas it differed signifi cantly (ANCOVA; mined based on the formula: P < 0.05) for Acropoma japonicum, Clupea pallasii palla- sii, Liparis tanakae, and Pseudorhombus pentophthalmus. W = aLb Lower b values were recorded in winter for C. pallasii pallasii, while A. japonicum in summer and winter, and where W is the wet weight [g], L is total length [cm], P. pentophthalmus were in summer. For L. tanakae, the b and a and b are the intercept and allometric coeffi cient, value was lower in autumn than other seasons. Seasonal respectively. Parameters a and b were estimated using differences in length–weight relations may be due to sea- a simple linear regression analysis of log-transformed sonal difference in the condition and/or sexual maturing data. Extreme outliers were removed before fi tting the of the specimens (Moutopoulos and Stergiou 2002, Froese linear regression (Froese et al. 2011). The 95% confi - 2006). dence intervals (CIs) of parameters a and b and the sta- Our LWR estimates were compared with the Bayesian tistical signifi cance level of r2 were estimated using SYS- estimates of LWRs in fi shes (Froese et al. 2014) and were TAT software (Systat, version 12. SPSS, USA). LWRs found to fall within the ranges expected for each species were summarized over all months. Seasonal LWRs, each or the genus and body shape where they belong. Thus, our spring (March–May), summer (June–August), autumn results represent the reasonable estimates of the LWRs for (September–November), and winter (December–Febru- the 29 species. The estimated parameters of LWRs in this ary), were also estimated for species where more than work should be used with some caution as the length rang- 30 specimens were collected for each season. Analysis es of the specimens available was limited by the selective of covariance (ANCOVA) was used to determine signif- characteristics of fi shing gear used. The majority of the icant differences in the slopes of length–weight regres- samples did not include the full size ranges of specimens sions between seasons. in the study area, especially early juveniles. Petrakis and Length–weight regressions were applied to 16 456 Stergiou (1995) suggested the use of LWRs only for spec- specimens of the 29 species from 24 families. Comparing imens of lengths within the length ranges used in the esti- with information in FishBase (Froese and Pauly 2015), our mation of the parameters. results provide new maximum lengths for 6 species: Ambly- In conclusion, this study provides LWRs of 29 fi shes chaeturichthys hexanema (15.6 cm TL), A. sciistius (9.2 cm caught off the south-eastern coast of Korea. Our results TL), Jaydia lineata (10.2 cm TL), Liparis tanakae (56.3 cm can be useful for sustainable utilization and management TL or 51.1 cm SL), Pseudorhombus pentophthalmus (26.8 of fi shery resources in the study area, and also for fi shery cm TL), and Zoarces gillii (45.1 cm TL). biologists/managers in Korea. The estimated parameters for the LWRs, along with the descriptive statistics by species, are given in Table ACKNOWLEDGEMENTS 1. All LWRs were highly signifi cant (P < 0.001), with We thank Dr Ki Moon Nam (Korea National Park Ser- r2 values > 0.938. The r2 values ranged from 0.938 vice, Korea) and Ja Hyuk Gu (commercial fi sherman) for for Acropoma japonicum to 0.993 for Liparis tana- assistance with our sampling. kae. The estimated b values ranged from 2.730 for Amblychaeturichthys hexanema to 3.440 for Sillago si- LWRs for 29 fi shes from south-eastern coast of Korea 429

3 0 0 1 0 56 82 74 90 90 87 2 988 978 970 974 993 986 960 976 989 .985 .981 Table 1 Table br 95% CIs of b aL = W g 2006 ab new records of maximum length. 2 95% CIs of a rst records of LWR for the species; fi 1 dence intervals; Length [cm] [g] Weight Min. Max. Min. Max. sh species on the south-eastern coast of East Sea, Korea durin 31 13.1 35.9 11.4 244.0 0.0048 0.0032–0.0071 3.044 2.915–3.172 0. 56 4.8 10.5 54 2.1 21.8 4.5 0.0137 0.0108–0.0193 10.7 3.116 49 2.964–3.267 0.7 11.3 0.973 94 12.3 45.1 73 0.0067 0.0050–0.0090 6.9 5.4 3.174 4.1 3.021–3.326 15.6 395.0 0.971 9.2 0.0040 2.5 0.0029–0.0054 3.038 0.5 2.949–3.127 23.8 0.0116 0.9 7.5 0.0085–0.0160 2.730 0.0061 2.598–2.862 0.0046–0.008 3.115 0.951 2.976–3.253 0.967 36 30 9.8 72 15.2 20.6 7.2 42.2 5.3 26.4 49.5 74.3 1279.9 3.1 0.0021 0.0135 0.0014–0.0033 0.0075–0.0242 256.3 3.440 3.073 3.285–3.596 0.0044 2.897–3.249 0.0036–0.0054 0.983 3.338 0 3.256–3.419 0.99 54 11.7 28.5 2.7 67.9 0.0009 0.0005–0.0014 3.346 3.179–3.507 0.97 n 117 117 5.7 18.1 1.9 75.9 0.0067 0.0057–0.0078 3.210 3.146–3.274 0.98 112 112 9.3 34.2 13.5 772.5 0.0236 0.0178–0.0311 2.884 2.795–2.973 0. 241 26.2 62.6 6.3 122.3 0.0001 0.0001–0.0002 3.357 3.247–3.468 0. 393 7.4 30.4 406 1.4 832 6.0 107.3 0.0047 56.3 9.1 0.0057–0.0057 2.857 29.1 2.2 2.783–2.931 1742.8 0.9 11.0 0.0098 0.0089–0.0108 346.1 282 3.065 0.0208 3.033–3.097151 0.0192–0.0241 2.871 6.2 0. 2.818–2.925 6.2 18.7 0. 17.4173 1.3 404 1.5 10.5 36.4 10.2 0.0071 20.6 26.8 0.0065–0.0078 26.8 2.912 0.0082 21.6 2.874–2.950 0.0067–0.0102 9.8 2.830 161.9 0.99 2.740–2.919 0.0140 208.7 0.0115–0.0169 0.963 0.0055 3.086 0.0045–0.0068 3.014–3.157 3.207 3.140–3.275 0 0. 472 5.0 57.4 1.1 3201.5 0.0199 0.0161–0.0245397 2.955 2.892–3.019 7.1 0. 34.5 2.6 362.9 0.0050 0.0045–0.0055 3.205 3.171–3.239231 0.9 135 9.5 8.0 44.5 19.7 8.7 3.3 958.1 0.0078 50.9 0.0069–0.0097 0.0043 3.136 0.0027–0.0066 3.092–3.180 3.175 3.009–3.341 0.9 0.96 101 14.9 164 49.5 7.7 4.4 15.9 212.1 0.0006 2.0 0.0004–0.0009 3.273 24.5 3.170–3.376 0.0029 0. 0.0021–0.0040 3.215 3.087–3.343 0.938 3368 3981 3.6 13.6 3.5 10.2 0.8 0.5 32.3 0.0129 0.0117–0.0141 17.5 3.036 0.0072 2.990–3.083 0.0063–0.0082 3.414 0.9 3.345–3.483 0.9 3947 6.7 28.7 1.7 171.4 0.0026 0.0025–0.0028 3.283 3.255–3.311 0. 1,2 2 1 1,2 1 1 1 cient of determination, CIs = confi 1 1 1 1 1 = coeffi 2 r 2 2

1,2 = slope, b Liparis tanakae Trichiurus lepturus Trichiurus Psenopsis anomala Amblychaeturichthys hexanema Sphyraena pinguis Lophius litulon Zeus faber rubripinnis Paracentropogon japonicum Acropoma Doederleinia berycoides semilineatus Ostorhinchus Callionymus valenciennei Amblychaeturichthys sciistius pentophthalmus Pseudorhombus Lepidotrigla guentheri Jaydia lineata Sillago sihama japonicus Trachurus Pagrus major Pennahia argentata gillii Zoarces Callionymus lunatus yokohamae Pseudopleuronectes Cynoglossus robustus Conger myriaster Coilia nasus Engraulis japonicus Clupea pallasii Coelorinchus multispinulosus = intercept, a Parameters of length–weight regression (LWR) for 29 fi Parameters of length–weight regression (LWR) Liparidae Trichiuridae Centrolophidae Gobiidae Sphyraenidae Triglidae Apogonidae Sillaginidae Carangidae Sparidae Sciaenidae Zoarcidae Callionymidae Pleuronectidae Cynoglossidae Clupeidae Order Family Species = number of individuals studied, n LophiiformesZeiformesScorpaeniformes Lophiidae Scorpaenidae Zeidae Perciformes Acropomatidae Pleuronectiformes Paralichthyida Anguilliformes Congridae Clupeiformes Engraulidae Gadiformes Macrouridae

430 Park and Huh 4 44 60 2 985 960 985 974 968 986 988 975 961 987 .932 .943 .987 .996 .972 Table 2 Table 0.983 0.991 0.994 0.982 0.912 0.975 0.945 0.984 0.989 8 0.920 2 0.982 47 0.987 0.984 9 0.975 5 0.957 br 17 0.928 95% CI of b aL 3.157 3.032–3.237 0.979 ast Sea, Korea during 2006 = W ab 1111 95% CI of a sh species which were abundant in the south-eastern coast of E Length [cm] [g] Weight Min. Max. Min. Max. n SpSu 112 3710 13.8 6.7 28.7 28.0 12.2 1.7 132.1 171.4 0.0031 0.0025 0.0017–0.0057 0.0023–0.0028 3.235 3.305 3.021–3.44 3.262–3.349 SpSu 160 51 7.4 7.7 29.3 28.6 1.4 1.9 101.6 0.0054 84.1 0.0060 0.0037–0.0079 0.0035–0.0102 2.803 2.790 2.658–2.948 2.595–2.985 0 0.9 SpSu 979 544 3.6 5.1 10.5 12.4 0.8 1.6 16.8 21.4 0.0142 0.0156 0.0122–0.0166 0.0104–0.0234 3.008 2.912 2.927–3.088 2.772–3.103 0. 0. SpSp 64 Su 5.0 245 104 57.4 6.0 13.7 1.1 56.3 40.6 3201.5 2.2 24.5 0.0074 1742.8 786.9 0.0064–0.0086 0.0100 0.0094 3.219 0.0086–0.0116 0.0070–0.0127 3.172–3.265 3.050 3.091 0 2.999–3.101 2.998–3.18 SpSu 1648 34 4.1 6.2 10.0 9.0 0.9 3.6 17.5 0.0055 12.8 0.0072 0.0044–0.0068 0.0037–0.0139 3.526 3.408–3.644 3.371 3.049–3.693 0 0.93 SpSu 97 35 8.0 11.0 34.5 29.5 3.9 11.4 362.9 271.3 0.0058 0.0047 0.0047–0.0072 0.0035–0.0062 3.146 3.224 3.068–3.223 3.133–3.316 0. SpSu 143 26 6.2 9.4 15.5 18.7 1.3 4.9 22.4 36.4 0.0070 0.0069 0.0060–0.0082 0.0033–0.0146 2.918 2.954 2.845–2.991 2.675–3.234 0. 0.9 SpSu 120 59 12.0 14.4 26.8 25.3 15.1 24.1 208.7 166.1 0.0064 0.0081 0.0045–0.0092 0.0037–0.0178 3.169 3.077 3.053–3.28 2.822–3.332 SpSu 99 42 9.5 13.0 32.1 35.0 8.7 27.4 399.5 557.7 0.0085 0.0091 0.0070–0.0101 0.0063–0.0132 3.103 3.095 3.040–3.167 2.970–3.220 0. Wi 104 12.8 25.9 11.4 121.5 0.0038 0.0029–0.0051 3.141 3.040–3.24 Wi 174 7.4 30.4 1.8 107.3 0.0049 0.0034–0.0071 2.836 2.705–2.967 0 Wi 51 21.3 49.3 193.4 1563.4 0.0174 0.0085–0.0360 3.003 2.790–3.2 Wi 1176 4.6 10.6 1.1 19.9 0.0082 0.0070–0.0096 3.278 3.191–3.364 0 Wi 1556 4.3 9.2 1.0 14.3 0.0100 0.0078–0.0129 3.216 3.070–3.362 0. Wi 150 7.1 19.3 2.6 60.7 0.0058 0.0048–0.0070 3.136 3.063–3.209 0. Wi 102 9.3 18.3 5.0 35.0 0.0083 0.0053–0.0135 2.843 2.664–3.023 0. Wi 73 14.8 25.1 28.4 173.2 0.0097 0.0058–0.0163 3.023 2.851–3.195 Wi 38 11.6 44.5 16.3 958.1 0.0078 0.0057–0.0107 3.130 3.022–3.237 Au 669 4.4 13.6 1.1 32.3 0.0132 0.0112–0.0154 3.022 2.946–3.097 0. Au 57 23.1 41.8 167.2 941.6 0.0158 0.0101–0.0248 2.938 2.809–3.06 Au 14 16.7 47.1 58.8 2330.0 0.0147 0.0092–0.0236 3.018 2.877–3.15 Au 743 4.0 10.2 1.0 17.6 0.0090 0.0068–0.0119 3.276 3.120–3.433 0. Au 115 10.4 30.5 9.3 285.6 0.0048 0.0039–0.0061 3.225 3.144–3.305 Au 152 10.2 24.4 9.8 145.5 0.0052 0.0039–0.0068 3.218 3.122–3.314 Au 52 14.1 35.2 32.1 497.1 0.0076 0.0052–0.0 Species Season Parameters of seasonal length–weight regression (LWR) for 10 fi Parameters of seasonal length–weight regression (LWR) Clupea pallasii Coelorinchus multispinulosus Acropoma japonicum Acropoma Lophius litulon Liparis tanakae Jaydia lineata Trachurus japonicus Trachurus Callionymus lunatus Pseudorhombus pentophthalmus Pseudorhombus Pseudopleuronectes yokohamae Pseudopleuronectes Sp = spring, Su summer, Au autumn, Wi winter. LWRs for 29 fi shes from south-eastern coast of Korea 431

REFERENCES Moutopoulos D.K., Stergiou K.I. 2002. Length–weight Baeck G.W., Jeong J.M., Kim H.J., Huh S.H., Park and length–length relationships of fi sh species of the J.M. 2014. Length–weight and length–length relation- Aegean Sea (Greece). Journal of Applied Ichthyology ships for four species of righteye fl ounder (Pleuronec- 18 (3): 200–203. tidae) on the south coast of Korea. Journal of Applied DOI: 10.1046/j.1439-0426.2002.00281.x Ichthyology 30 (1): 204–205. Petrakis G., Stergiou K.I. 1995. Weight–length relation- DOI: 10.1111/jai.12310 ships for 33 fi sh species in Greek waters. Fisheries Re- Baeck G.W., Jeong J.M., Yeo Y.M., Huh S.-H., Park search 21 (3–4): 465–469. J.M. 2012. Length–weight and length–length relation- DOI: 10.1016/0165-7836(94)00294-7 ships for 10 species of scorpionfi shes (Scorpaenidae) Ricker W.E. 1968. Methods for assessment of fi sh pro- on the south coast of Korea. Journal of Applied Ichthy- duction in freshwaters. Blackwell Scientifi c Publica- ology 28 (4): 677–679. tions, Oxford and Edingburgh, UK. DOI: 10.1111/j.1439-0426.2012.01956.x Suh Y.S., Jang L.H., Hwang J.D. 2001. [Temporal and Froese R. 2006. Cube law, condition factor and weight– spatial variations of the cold waters occurring in the length relationships: History, meta-analysis and rec- eastern coast of the Korean Peninsula in summer sea- ommendations. Journal of Applied Ichthyology 22 (4): son.] Journal of the Korean Fisheries Society 34 (5): 241–253. 435–444. [In Korean.] DOI: 10.1111/j.1439-0426.2006.00805.x Wang X., Xue Y., Ren Y. 2013. Length–weight relation- Froese R., Pauly D. (eds.) 2015. FishBase. [Version ships of 43 fi sh species from Haizhou Bay, central 03/2015] www.fi shbase.org Yellow Sea. Journal of Applied Ichthyology 29 (5): Froese R., Thorson J.T., Reyes R.B. 2014. A Bayesian 1183–1187. approach for estimating length–weight relationships in DOI: 10.1111/jai.12200 fi shes. Journal of Applied Ichthyology 30 (1): 78–85. Xue Y., Ren Y., Xu B., Mei C., Chen X., Zan X. 2011. DOI: 10.1111/jai.12299 Length–weight relationships of fi sh species caught by Froese R., Tsikliras A.C., Stergiou K.I. 2011. Editorial bottom trawl in Jiaozhou Bay, China. Journal of Ap- note on weight–length relations of fi shes. Acta Ichthy- plied Ichthyology 27 (3): 949–954. ologica et Piscatoria 41 (4): 261–263. DOI: 10.1111/j.1439-0426.2010.01607.x DOI: 10.3750/AIP2011.41.4.0 Gonçalves J.M.S., Bentes L., Lino P.G., Ribeiro J., Received: 26 October 2015 Canário A.V.M., Erzini K. 1997. Weight–length re- Accepted: 9 December 2015 lationships for selected fi sh species of the small-scale Published electronically: 31 December 2015 demersal fi sheries of the south and south-west coast of Portugal. Fisheries Research 30 (3): 253–256. DOI: 10.1016/S0165-7836(96)00569-3